The Data Management and Biostatistics Core (DMBC or Core) provides the ADRC Projects and Core investigators with support in study design, sampling, data management, and statistical data analysis for presentation and publication. Complex modeling is best handled within the individual projects and can be seen as already well documented in each project. These modeling techniques were incorporated with the collaboration of the BC investigators. The specific aims of this core are to: 1) Provide data management skills and resources for Center projects; (2) Provide expertise in study design and statistical planning of research studies; Also plan and conduct statistical analyses as well as complex modeling for research projects which includes planning and conducting project data analyses for subsequent issues generated by the projects and data resulting from modeling within the projects; (3) participate in educational and training activities; and (4) collaborate fully with the National Alzheimer's Coordinating Center in providing them with timely and accurate data submissions. The statistical needs of the grant are in three primary areas: multivariate statistics, analysis of correlated data, and flexible regression modeling including non linear modeling. The DMBC will create new databases for all projects, participating in designing data collection instruments, coding shemes, training data collection personnel, computerizing data and developing and performing quality control functions. Traditionally data has been transmitted to the DMBC electronically via fax or file transfer or key entered by Core personnel. Data management and data entry support will be provided to all investigators affiliated with the ADRC. As a result of our modeling expertise, and should the opportunity present itself, the DMBC will provide methodological expertise to ADRC. It is our purpose to participate fully in Center activities including data presentations, Center Training activities including seminars, curriculum advise to fellows, and mentoring.